B. K. N. Wyllie & A. Scott — Plutonics of Garahal Hill. 541 



amphibole generally occurs round the diopside. It likewise seems 

 most probable that the alteration took place after consolidation. The 

 simplest explanation would seem to be that the change is due to the 

 oxidation of the ferrous oxide and consequent passage of the iron 

 from the basic radicles to the acid ones, to form ferrisilicates. The 

 chemical analyses, however, are against this, as the hornblendite 

 actually has less ferric iron than the diorite, while there is no reason 

 to suppose that the brown colour of araphiboles is due essentially to 

 ferric oxide : indeed, analyses of brown hornblende are recorded, 

 containing only a small percentage of ferric iron. That the green 

 hornblende is a metastable form is shown by the separation of the 

 minute crystals of the dark bands. Since these are too small to be 

 identified directly all that we can deduce respecting them is that 

 they are probably monoclinic hornblende or pyroxene, since the 

 rhombic modifications require special conditions such as would inhibit 

 their formation in the solid state. 



Two explanations suggest themselves: — 



1. The brown colour may be due to ultramicroscopic crystals of 

 the same nature as those of the feathery bands. As has been stated 

 above, the colour of pitchstones and other glasses is often due to 

 minute crystallites, which are too small to be resolved under the 

 microscope, and the only evidence of the presence of which is the 

 brown colour of the glass. Many pitchstones, too, show two genera- 

 tions of crystallites, one microscopic, the other sub-microscopic. 

 Hence it is plausible to suppose that the brown colour of the 

 hornblende has originated in this way, by the formation of in- 

 numerable crystallites of an order of magnitude lower than that of 

 the crystals of the dark bands. The only effects on the pleochroism 

 of the separation of such crystallites would be a uniform inci'ease in 

 the absorption, the difference for light waves traversing the crystal 

 in different directions remaining practically constant, and a change 

 in colour due to the total reflection of the light at the violet end of 

 the spectrum. 



2. On the other hand, the explanation may be that the metastable 

 green hornblende breaks up into two more stable minerals, the minute 

 crystallites of the dark bands and a brown hornblende. Analysis 

 No. Ylla was obtained by deducting from analysis No. VII the 

 amount of the various oxides which go to make up the other minerals 

 in the hornblendite, quartz, orthoclase, andesine, and biotite, the 

 amount of these present being calculated by the Eosiwal method. 

 As these are only present in small amount, the error in assuming their 

 composition as normal is negligible, and hence No. Vila probably 

 represents fairly closely the composition of the hornblende. Com- 

 paring this with analysis No. VI, it is seen that the amphibole of the 

 coarser varieties contains a much greater proportion of sesquioxides 

 than that of the finer types, as not only does more of the alumina in 

 the latter rocks form felspar, but the iron partly crystallizes as biotite. 

 In addition, the minute crystallites are colourless or nearly so, and 

 hence may be assumed to be salts of lime or magnesia, with the 

 result that their separation is accompanied by an enrichment of 

 the residual material in ' total ' iron. Thus we would have the 



